A multicellular organism is composed of many types of cells performing specialized functions. Cells communicate with each other for the organism to function as a whole. They do so at many levels and by various mechanisms. Cell identity is determined by the proteins that are synthesized within the cell. Therefore, regulation of gene expression, especially transcription, is the key mechanism of controlling cell growth and differentiation. To control transcription in response to extracellular stimuli originating from other cells or the surrounding environment, signals from outside the cell are transmitted to the transcription machinery inside the nucleus via a variety of signaling molecules including receptors, protein kinases and phosphatases and adapters, which form networks known as signal transduction pathways, or STPs (Boulikas, 1995, Crit. Rev. Euk. Gene Exp. 5: 1-77; Hunter et al., 1992, Cell 70: 357-387; Karin and Hunter, 1995, Curr. Biol. 5, 747-757; Treisman, 1996, Current Opinion in Cell Biol. 8: 205-215). These signaling molecules are also of interest to medical science and to the pharmaceutical industry because malfunctions of these molecular pathways are the cause of many human diseases, including cancers, genetic disorders and immune diseases. Therefore, many of the intracellular signaling molecules are targets for drug intervention and targets for drug screening.
Among the best characterized STPs are the cAMP-dependent protein kinase (PKA) pathway and the mitogen-activated protein kinase (MAPK) pathways mediating signals from growth factors (e.g. EGF and NGF) and cellular stress such as heat, UV, oxidative stresses and protein synthesis inhibitors (FIG. 1). These signaling pathways receive diverse upstream signals and cause distinct downstream changes. One common feature of these pathways is that they all rely on the nuclear translocation of an activated protein kinase (e.g. MAPK or PKA) to transmit signals from outside the cell to the transcription machinery in the cell nucleus (FIG. 1). When activated by upstream signals, signal transduction kinases specific to a given pathway (e.g., MAPK or PKA) translocate into the nucleus and phosphorylate critical residues thereby activating the transactivating activity of a specific transcription factor(s), and thus converting an extracellular signal into a specific transcriptional response in the nucleus (FIG. 1). The activation status of the transcription factors therefore reflects the activation status of the respective kinases and upstream signaling molecules along the pathway.
Since the discovery of the first member of the MAPK family (Ray and Sturgill, 1987, Proc. Natl. Acad. USA 84: 1502-1506), more than one hundred MAPK family members have been cloned (Kultz, 1998, J. Mol. Evol. 46: 571-588). These signaling pathways all use a three-component protein kinase cascade consisting of MAPK/MAPK kinase/MAPK kinase kinase but receive diverse upstream signals and cause distinct downstream changes. In the budding yeast S. cerevisiae, at least five distinct MAP kinase pathways have been identified to function in mating (Fus3/Kss1), cell wall biosynthesis (Mpk1), osmosensing (Hog1), sporulation (SmK1) and pseudohyphal development and invasive growth pathway (Levin and Errede, 1995, Current Opin. Cell Biol. 7: 197-202; Waskiewicz and Cooper, 1995, Curr. Opin. Cell. Biol. 7, 798-805). In mammalian cells, over 12 MAPKs have been cloned and characterized (Kultz, 1998, supra; Waskiewicz, 1995, supra). Among the transcription factors activated by MAPK pathways is CHOP/GADD153, which will be referred to herein as CHOP (Wang and Ron, 1996, Science 272: 1347-1349). CHOP is activated by p38 MAPK, which is in turn activated by the MAPKK MEK3. CHOP activity thus reflects the avtivationof those factors upstream of CHOP in the activation pathway.
Transcriptional activator proteins have been found to be modular in nature, very often comprised of linked domains that retain their respective functions when separated from the remainder of the protein. This modular nature of transcription factors was originally demonstrated in the yeast GAL4 transactivator, and has subsequently been found in a wide variety of transcription factors wherein the activation domain (AD) and DNA binding domain (DBD) may be structurally and functionally separated (Ma and Ptashne, 1987, Cell 48: 847-853). An important consequence of the modular nature of transcription factors is that the isolated functional domains may confer the specific DNA binding or transactivating activity of a given transcription factor upon an unrelated fusion partner.
Fusion transactivators consisting of the DNA-binding domain of yeast GAL4 protein (amino acid residues 1-92 or 1-147; Sadowski and Ptashne, 1989, Nucl. Acids Res. 17: 7539) or E. coli LexA (residues 1-87; Thliveris and Mount, 1992, Proc. Natl. Acad. Sci. USA 89: 4500-4504) and the activation domains of transcription activators from higher eukaryotes have been used in the literature as sensors for specific pathways in transient transfection assays (FIG. 2; Xu et al., 1997, Strategies 10: 1-3; Xu et al., 1997, Strategies 10: 79-80; Xu et al., 1997, Strategies 10: 81-83; Sanchez et al., 1998, Strategies 11: 52-53; Baichwal and Tjian, 1990, Cell, 63: 815-826; Enslen et al., 1996, Proc. Natl. Acad. Sci. USA. 93: 10803-10808; Hill et al., 1993, Cell, 73: 395-406; Lin et al, 1995, Science, 268: 286-289; Livingstone et al, 1995, EMBO J., 14: 1785-1797; Marais et al., 1993, Cell 73: 381-393; Minden, et al., 1995, Cell, 81: 1147-1157; Price et al., 1995, EMBO J, 14: 2589-2601; Smeal et al., 1994, EMBO J., 13: 6006-6010; Wang and Ron, 1996, Science, 272: 1347-1349). A trans-reporting system of this kind includes a fusion transactivator plasmid that expresses a fusion protein consisting of the activation domain of a pathway-specific transcription factor and the DNA binding domain of a sequence-specific DNA binding factor. The transactivation moiety of the fusion transactivator is phosphorylated and activated by kinases specific to that activation pathway. The activity of the fusion activators, therefore, reflects the in vivo activation of the specific kinases and the corresponding signal transduction pathways. The DNA-binding domain moiety enables the fusion activator to bind one or more copies of the binding element (e.g., that from GAL4, LexA or other sequence-specific DNA binding protein) situated upstream of the reporter gene in a separate reporter vector (FIG. 2). Expression (or activity) levels of reporter gene product reflect the activation status of the signaling pathway. Therefore, the effects of a gene product or an extracellular stimulus such as a growth factor or UV irradiation can be monitored by simple and sensitive reporter assays.
The systems described above are transient transfection assay systems. Introduction of plasmid DNA into mammalian cells by transfection is still largely a trial and error process. Transfection efficiency fluctuates from cell to cell and from experiment to experiment, which can give rise to inconsistent assay results. Some applications of pathway-specific signal transduction systems, especially those involving a high volume of samples such as drug screening applications, demand more convenient format and more consistent results. There is a need in the art for pathway-specific signal transduction assay systems that provide consistent assay results.
GAL4 fusion proteins have been used to test protein:protein interactions, to study chromatin structure and function and to serve as inducible transcription factors for protein expression and the measurement of the biological activities of steroid hormones (Jausons-Loffreda et al., 1994, J. Biolumin. Chemilumin. 9: 217-221; Braselmann et al., 1993, Proc. Natl. Acad. Sci. USA 90: 1657-1661; Louvion et al., 1993, Gene 131: 129-134; Dang et al., 1991, Mol. Cell. Biol. 11, 945-962). The steroid hormone studies investigated the single component steroid hormone receptor pathway, in which the receptor itself translocates to the nucleus and transactivates gene expression after binding the hormone. Fusion proteins of this kind have not been used, however, to establish stable reporter systems responsive to multicomponent signal transduction cascade pathways such as the MAPK pathway. There is a need in the art for systems that re-create the natural cellular environment for activation involving multicomponent signal transduction cascades.
There is also a need in the art for pathway-specific signal transduction assay systems that do not require transfection, whether transient or otherwise.